A SCOPING STUDY ON
THE EPHEMEROPTERA
OF SOUTHERN CHALK
STREAMS
A Report to the Envinulinent Agency
.1 F Wright .1 II Blackburn 121 NI Gunn K I. Symes •1 Bowker
Institute of htshuater Ecology River Laboratory East Stoke W.NI2E11.1N1 121120 611B
A pril 1998 =I IM 61•1•11111•1=10110110111100 MEN Scoping Study on the Ephemeroptera of Southern Chalk Streams
J F Wright, J H Blackburn, R J IVIGunn, K L Symes & J Bowker
Research contractor: Institute of Freshwater Ecology
Environment Agency Rivers House Sunrise Business Park Higher Shaftesbury Road Blandford Forum DTI 1 8ST Publishing Organisation Environment Agency Rivers House Sunrise Business Park Higher Shaftesbury Road Blandford Forum DT11 8ST
Tel 01258 456080 Fax: 01258 455998
C) Environment Agency 1998
All rights reserved. No part of this document may be produced, stored in a retrieval system or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise without the prior permission of the Environment Agency.
The views expressed in this document are not necessarily those of the Environment Agency. Its officers servant or agents accept no liability whatsoever for any loss or damage arising from the interpretation or use of the information, or reliance upon views contained herein.
Dissemination Status Internal: Released to South-West, Southern, Thames and Anglian Region. External: Restricted
Statement of Use This short scientific review was commissioned to further a 'LEAP' action to investigate factors which potentially could contribute to an alleged decline in Ephemeroptera within the Avon catchment. The review has relevance to other chalk systems within the UK and should significantly contribute to the Agency's knowledge, enabling future monitoring to be focused on specific aspects which influence this group.
Research Contractor This document was produced under the Technical Services Agreement by:
Institute of Freshwater Ecology River Laboratory East Stoke Wareham Dorset BH20 6BB
Tel: 01929 462314 Fax. 01929 462180
Environment Agency Project Manager Mr A. Frake, Environment Agency, South West Region. ACKNOWLEDGEMENTS
The authors would like to thank the Environment Agency, South West Region for initiating this study. In particular, we thank Dr. G.P.Green and Dr. D.A.Cooling for helpful discussions at the outset and Mr. A Frake for his advice as Nominated Officer to the project. Funding for the report was provided under the Technical Services Agreement between the Environment Agency and the Institute of Freshwater Ecology.
INTELLECTUAL PROPERTY RIGHTS
CONFIDENTIALITY STATEMENT
'In accordance with our normal practice, this report is for the use only of the party to whom it is addressed, and no responsibility is accepted to any third party for the whole or any part of its contentg Neither the whole nor any part of this report or any reference thereto may be included in any published document, circular or statement, nor published or referred to in any way without our written approval of the form and context in which it may appear.' WIIMMIIINIMUNIIMMIIMMIMMININIMOINIII0
CONTENTS Page
Acknowledgmen ts
List of Tables
List of Figures
List of Appendices
Executive Summary vii
Key Words viii
I. Introduction 1.1 Background to the Study 1 1.2 The British Ephemeroptera (mayflies) 1
2. The Ephemeroptera of southern chalk streams 3 2.1 Species characteristic of southern chalk streams 3 2.2 Seasonal occurrence of larvae 5 2.3 Species richness at chalk stream sites 6 2.4 Longitudinal occurrence of species 8 2.5 Abundance of families at chalk stream sites 9
3. Notes on the life cycles of each species 13
4. Factors affecting the distribution and abundance of Ephemeroptera 17 4.1 Habitats, habits and feeding behaviour of the larvae 17 4.2 The impact of discharge regime 19 4.3 The impact of pollution 21 4.4 Factors affecting the terrestrial (adult) phase 23
5. References 25
Appendices 29 1•11-1•11•11MOIIININIIIIIIIMIN•11•1111•10 a IIMINI LIST OF TABLES Page
21 Listing of the mayflies recorded at 42 chalk stream sites, together with 4 I the numbcr of sites at which each taxon was found. Data from RIVPACS samples (three seasons samples combined)
I 2.2 Frequency of occurrence of mayfly larvae in 42 chalk stream sites 6 by season
I 2.3 Longitudinal occurrence of mayfly larvae at 42 chalk stream sites 9
3.1 Information on the flight periods and life cycles of mayflies abstracted 14 I from Elliott & Humpesch (1983) and Elliott eta (1988), together with additional observations made on southcrn chalk streams
1 4 1 Running-water habitats (from Wright et al. 1993), together with 17 habits and feeding behaviour (from Elliott c/at 1988) for the I mayflies found in southern chalk streams
I LIST OF FIGURES
2 I Number of species of mayfly larvae recorded at 42 chalk stream sites 7
22 Mayfly species richness for 42 chalk stream sites in relation to location 7 downstream
2.3 Log. category abundance data for each mayfly family at 42 chalk stream I 1 sites in spring, summer and autumn.
LIST OF APPENDICES
I Appendix 1. Specification for the scoping study 29
Appendix 2. The 42 southern chalk stream sites for which RIVPACS samples 30 I have been obtained.
Appendix 3. Mayflies found at each of the 42 chalk stream sites Data for the 31 1 three seasons are combined (1 indicates taxon occurrence in a single season, whereas 2 (or 3) indicates presence in two (or 3) seasons.
I
1 MINIMIINIIIMMINIIIIIMMINIIIIIIMISM EXECUTIVE SUMMARY
This scoping study was undertaken at the request of the Environment Agency, following expressions of concern from fishermen that some species of Ephemeroptera (mayflies) appear to have declined on the R.Wylye and on the upper reaches of the Hampshire Avon. The purpose of this scoping study is to provide a short scientific review of the mayfly fauna of southern chalk streams.
• Forty nine species of mayflies are known to occur in Britain. The status of two additional species is uncertain. An appraisal of the R1VPACS III data-base yielded 42 high quality sites on southern chalk streams. Examination of the species lists for these sites confirmed that over 50% of the British species occur on southern chalk streams. Information is provided on the seasonal occurrence of the different species and their frequency of occurrence at the 42 sites.
Between four and fifteen species of mayflies were recorded per site, with an arithmetic mean of 9.12 per site. The most species-rich sites (12 or more species) occurred between 10 and 50, km from stream source. They included sites on such well known rivers as the R. Frome (Dorset), R. lichen, R. Lambourn, Moors River (Dorset) and R.Test. The study includes information on the longitudinal occurrence of each species.
The R1VPACS III data-set also includes information on the log, categories of abundance of each family of mayfly in spring, summer and autumn, based on the standard R1VPACS pond-net samples. This provides a useful general guide on whether a given family is scarcc, common or abundant in a given season. Members of the Baetidae were found to be important components of the mayfly fauna in spring, summer and autumn, but were particularly abundant at most sites in summer. Ephemerellidae were also very abundant at most sites in summer.
Notes are provided on the flight period and life cycles of each species, as far as these are known The life cycles of some species show flexibility and some members of the Baetidae are capable of having several generations in a year. Hence, interpretation of the life cycle can pose problems
Information on the habitats, habits and feeding behaviour of the larvae demonstrates that a wide variety of different habitats and food resources are exploited by mayfly larvae.
A consideration of the potential impact of flow regime on larval populations suggests that if changes in the flow regime affect the habitats and food resources within a stream, then there are likely to be consequences for the mayfly fauna. However, different members of the mayfly fauna may respond in different ways, given their various adaptations.
Low flows arc of particular interest at present, and several short term studies on chalk streams, together with one long-term study on the R.Lambourn in the 1970's, are starting to provide scientific data on the way in which different components of the
VII mayfly fauna appear to respond to the discharge regime. Some provisional findings for the Baetidae, Ephemerellidae and Ephemeridae are given below.
A nine-year study on the R. Lambourn provided evidence that in early June, densities of larvae in the Baetidae (which may have included up to six species) were low in drought years (-500 ni2) compared with some years with high discharge when densities were —5,000 1112.
Densities of larvae in the Ephemerellidae (Ephemerella ignita only) were higher in the year following a drought.
Densities of larvae in the Ephemeridae (Ephemera danica only) did not appear to be related to the prevailing discharge regime in the R. Lambourn. Instead, cool damp conditions during the period of flight activity in May/June of one year resulted in very low larval densities of the next generation in December of that year. Over the next three generations (at intervals of two years), larval densities increased progressively in the December following oviposition.
Studies commissioned by the Environment Agency on a number of chalk stream sites first examined in the 1970's may soon provide further information on the response of mayflies to the flow regime.
Some chalk streams may be subject to organic pollution as a result of sewage effluent, fish farm effluent or farm wastes. Low levels of organic enrichment may not have adverse effects on the mayfly fauna, but more severe cases of organic pollution will result in the progressive elimination of the more sensitive species. Environment Agency biologists are well-versed in the detection of organic pollution using standard methodologies.
Although the time spent in the aerial phase of the life cycle is only a small fraction of the total life span, it is crucial for successful reproduction and dispersal. Despite predation and parasites, the subimagos and imagos normally find suitable shelter and terrain markers for swarming, thereby ensuring that mating and subsequent oviposition is successful. However, as previously indicated for Ephemera danica, there is circumstantial evidence that inclement weather is capable of disrupting this critical phase of the life cycle.
KEY WORDS
Chalk stream; Ephemeroptera; mayflies; distribution; low flows; pollution I. INTRODUCTION
1.1. Background to the study
The need for this scoping study was identified during a meeting between staff from the Environment Agency (EA) and the Institute of Freshwater Ecology (IFE) which took place at the River Laboratory in late August 1997. EA staff indicated that there had been expressions of concern from fishermen, alleging a decline in Ephemeroptera (mayflies) within some tributaries of the Hampshire Avon (R.Wylye & Upper Avon). In particular, concern focused on poor hatches of mayflies in the family Baetidae.
As a result of the initial meeting, the ME was asked to develop some proposals under three separate headings.
A short scientific review of the mayfly fauna of southern chalk streams.
An appraisal of a very limited data-set held by the WE on the mayfly larvae of one or two sites on the R Wylye and Upper Avon for the 1970's, to determine whether there was merit in repeating these surveys.
An outline proposal for the long-term monitoring of chalk stream mayflies, with particular regard to the emergent stage of the life cycle.
Following a consideration of these outline proposals from the IFE, the EA requested separate quotations for items a). and c). Funding for a short scientific review of the mayfly fauna of southern chalk streams (item a) was approved in November 1997 and forms the subject of the present report. The agreed specification for the review is given in Appendix 1.
1.2. The British Ephemeroptera (mayflies)
Worldwide, the insect order Ephemeroptera is quite small, with something in excess of 2,000 known species in approximately 200 genera and 19 families (Brittain, 1982). Until recently, the British list included 48 species in eighteen genera and just eight families (Elliott, Humpesch & Macan, 1988). However, two of the species on this list, Arthroplea congener Bengtsson and Heinagenia longicauda (Stephens), have not been recorded since the 1920's/1930's and may no longer occur in the British Isles (Bratton, 1990). Refer to Elliott, Humpesch & Macan (1988) for the checklist of 48 British species.
In the last few years, examination of specimens from many streams and rivers in Great Britain has revealed three additional species previously recognised on mainland Europe, but unknown in Britain. These are Caenis pseuclorivulonnn Kieffermuller (Gunn and Blackburn, 1997) and Cactus heskidensis Sowa (Gunn and Blackburn (in press)) in the family Caenidae and Electrogena affinis (Eaton) (Blackburn, Gunn and Hammett (in press)) in the Heptageniidae. Thus, at present, 49 species of mayflies are known to occur in Britain with two additional species of uncertain status • OM NM 01 INSIMINNOWINIMMINNEINIIIIMMINNIMI 2. THE EPHEMEROPTERA OF SOUTHERN CHALK STREAMS
2.1. Species characteristic of southern chalk streams
The RIVPACS III data-set, which includes species-level data for many chalk stream sites of high quality in southern England, was used to determine which species occurred in chalk streams. For the purposes of this scoping study, the EA nominated officer and IFE staff agreed that 'southern' chalk streams would be defined as occurring in three EA regions (Thames, Southern and South-West region) and that whereas the source waters would always be derived from chalk aquifers, sites would continue to be defined as chalk stream sites when the course of the river progressed over different geology (e.g. tertiary gravels). This protocol generated a total of 42 chalk stream sites which are listed in Appendix 2. In turn, these sites yielded 26 mayfly taxa (mainly species, but including two genera and two species 'groups') in six families. There were no records for two families (Siphlonuridae and Potamanthidae).
A list of the mayflies which were characteristic of chalk stream sites, together with their frequency of occurrence at the 42 sites is given in Table 2.1. The Baetidae, represented by 12 taxa in four genera, was by far the most taxon-rich family in southern chalk streams. All taxa were identified to species with the exception of the Bactis scambus group, which comprised both B. scambus Eaton and B.fuscaffis (Linnaeus). These two species, which are difficult to separate as larvae, were always treated as a 'species group' in the RIVPACS project. Within the genus Baths, one species (B.digitatus), is known from a very limited number of sites in Britain (including non-chalk stream sites) but others, such as B.rhodani, B.vernus and B.scambus group occurred in a majority of the chalk stream sites and normally formed an important element of the fauna. Additional species, including B.muticus and B.niger were frequent and B. atrebatinus and B. buceratus were also encountered quite frequently. Of the remaining species in the Baetidae, Centroptilum luteolum occurred in over half of the chalk stream sites, but C. pennulatum, Procloeon bifidum and Cloeon dipterum were less commonly encountered.
Members of the family Heptageniidae, are usually associated with fast-flowing upland streams, but one species, Ileptagenia sulphurea, is characteristic of chalk stream sites. Two other genera were also represented, but were infrequent in the data-set. They were Rhithrogena sp.(either R. germanica or R.semicolorata) and Ecdyonurus sp. (four species in this genus). Within the RIVPACS project, an early decision was made not to attempt to distinguish the larvae of the different species in these two genera due to inherent problems with small specimens Hence, the results are presented at generic level
The family Leptophlebiidae, includes three genera and all were represented in the chalk stream data-set. However, Leptophlebia vespertina, which is more characteristic of still waters, was recorded only once and Habrophlebia fusca was only recorded at two sites. All three species in the genus Paraleptophlebia were present and Paraleptophleffia suhmarginata was found at 20 of the 42 sites. In contrast, P.cincla
3 was only recorded twice and /' werneri, a rare species with Red Data Book 3 status (Bratton, 1990), which is most characteristic of winterbourne streams (intermittent chalk streams), was found at a single site.
The Ephemeridae includes three British species but only Ephemera danica, the fisherman's mayfly, was a common and characteristic species of southern chalk streams.
Table 2.1 Listing of the mayflies recorded at 42 chalk stream sites, together with the number of sites at which each taxon was found. Data from RIVPACS samples (three seasons samples combined)
Family S eciesNo. of Occurrences
Bactidac Baetis atrebatinus EatonI I &ens buceratus Eaton10 Baths chgatatus Bengtsson1 Baths muticus (L.)17 Baebs mger (L.)I 7 Bae rhodatu (Pictet)41 Baths vernus Curtis37 Baths scambus group'30 Centroptilutn luteolum (Muller)24 Centroptilurn pennulatum Eaton6 Cloeon dipterum (L.)2 Procloeon bifidum (Bengtsson)3
Heptagenlidae Rhathrogena sp 5 lieptagenta stdphurea (Muller) 24 Ecdyonurus sp. 4
Leptophlebiidae Leptophlebin vespertina (L ) Paraleptophlebta cmcta (Retzaus) 2 Paredeptophlebia submargtnata (Stephens) 20 Paraleptophlebta.werneri Ulmer 1 Habrophlebta fusca (Curtis) 2
Ephemeridac Ephemera dantca Muller 31
Ephcmcrellidae Ephemerella ignita (Poda) 42
Caenidac3 Brachycercus harrisella Curtis 2 Caenis horaria (L.) 1 Cactus rivutorum Eaton 24 Caenis luctuosa group2 25
Includes Baths scambus Eaton and &ens fuscams (Linnaeus)
2 Includes Caenis lucluosa (Burmeister) and Cactus macrura Stephens
3 Recently, Caems pusdla was identified amongst Caenids in samples from the Test, ltchen & Hants Avon. However, the data-base has not yet been updated.
4 The family Ephemerellidae includes two species, of which just one, Ephemerella ignita, known to fishermen as the blue-winged olive, was present at every site.
Finally, members of the Caenidae, often referred to as the Angler's Curse, were represented by four taxa. Two of them (Brachycercus harrtsella and Caenis horaria) were rarely recorded, being more characteristic of other types of flowing and standing waters. However, Caents rivulorum and the Caenis luctuosa group (which includes both C.luctuosa (Burmeister) and C. macrura Stephens, which are difficult to separate as larvae) were both encountered at more than half of the chalk stream sites in the data-set.
A full listing of the mayflies recorded at each of the 42 chalk stream sites is given in Appendix 3. This list combines the records for the three seasons (spring, summer and autumn) in which the RIVPACS samples were collected, such that 1 indicates that a given taxon occurred in a single season, whereas 2 (or 3) indicates presence in two (or 3) seasons
2.2. Seasonal occurrence of larvae
Each RIVPACS sample was collected by pond-net and involved sampling all available habitats in proportion to their occurrence over a period of 3 minutes in each of spring, summer and autumn. These seasons were broadly defined due to the scale of the sampling operation. They were spring (March - May), summer (June - August) and autumn (September - November).
The frequency of occurrence of each species by season is given in Table 2.2. The results provide a broad indication of the season or seasons in which the larvae of a given species may be found and whether they are common, frequent or rare in a chalk stream.
However, it would be unwise to over-interpret the information in the table without prior knowledge of the life cycles of the individual species (See Section 3 of this report). For example, both Baths rhodani and Ephemera danica occur at many sites in spring, summer and autumn, but whereas in chalk streams, B. rhodani appears to have several generations per year (multivoltinc), E.danica has one generation every two years (semivoltine). Again, B.muticus and B. niger overwinter as larvae and are most frequent in spring, whereas B.vernus and B.scambus group overwinter in the egg stage and are most frequent in summer.
In cases where there are few records for a species, it is even more important to be aware of the life cycle, because a single record in a given season may have very little or alternatively, considerable significance. An example of the latter is the single record of Paraleptophlebta werneri in spring. This species is characteristic of calcareous streams, and in particular winterbournes. When present in winterbournes, the larvae must complete their larval stage in spring and emerge before the stream dries up in summer. Resistant eggs in the stream bed then survive the period of drought through the autumn and early winter. Table 2.2 Frequency of occurrence of mayfly larvae in 42 chalk stream sites by season.
No. of Occurrences S ecics Sprin Summer Autumn
Thetis atrebatinus 4 6 6 Baetis buceratus 7 5 2 Baetis digitatus 0 1 0 Baetis muticus 15 11 5 Bactis niger 14 7 5 Bactis rhodani 35 33 34 Baetis vernus 14 33 27 Baetis scambus group 8 28 19 Centroptilum luteolum 13 13 11 Centroptilum pennulatum 0 5 3 Cloeon dipterum 0 1 1 Proclaeon bifidum 0 3 0
Rhithrogena sp. 5 2 2 Heptagenia sulphurea 21 12 18 Ecdyonurus sp. 1 3 2
Leptophlebia vespertino 1 0 0 Paraleptophlebia cincta 0 2 0 Parateptophlebia submarginata 10 3 17 Paraleptophlebia werneri 1 0 0 Hobrophlebia fusca 1 1 1
Ephemera danica 28 23 26
Ephemerella ignita 25 41 28
Brachycercus harrisella 0 2 0 Caenis horaria 0 0 1 Coenis rivulorum 18 5 9 Caenis luctuosa group 9 18 11
2.3. Species richness at chalk stream sites
The number of mayfly taxa recorded in the combined seasons RIVPACS samples at the 42 sites varied from 4 to 15 with an arithmetic mean of 9.12 taxa per site (Fig.2.1). These numbers should not be regarded as the full compliment of species occurring at each site, but probably give a reasonable indication of site richness. For example, the three 3-minute RIVPACS samples for the R.Lambourn at Bagnor yielded 12 species. A very comprehensive sampling programme undertaken throughout the 1970's yielded just two additional species.
In view of the variation in the number of species found at different sites, there is a necd to consider whether mayfly richness has a tendency to vary with location downstream
6 In •
7
6
S .
4
3 •
2
—t—t—tllt ui 1 2 3 4 5 6 7 8 9 10 II 12 13 14 15 16 17I1h18 19 20 NumberofSpecies
Figure 2 I Number of species of mayfly larvae recorded at 42 chalk stream sites
Figure 2.2 presents site species richness in relation to distance downstream. It is apparent that sites with relatively low richness (say, six or fewer species) are most frequently encountered at locations less than 20km or over 75km from source.
18
16 Is 211 14 7• 40 i • 65 7. 2k.., 12 85 9* •• • t.g 10 • • se• a • • • • •••• 1 8 es • E • Z 6 es a • a • • 4 le 2
0 I i 1 1 1 I 1 I
0 10 20 30 40 50 60 70 80 90 100
Kin from source
I Moom/Crane at Romlbrd Bridge 6 lichen dis Chickenhall SDW 2 Frome at Moreton 7 Teal at Skidmore 3 lichen at lichen St Cross Larnhoum at l3agnor 4 Test at Lower Brook 9 lichen at Otterhourne Water Works 5 Prone at East Stoke
Figure 2.2 Mayfly species richness for 42 chalk stream sites in relation to location downstream.
7 In contrast, sites with twelve or more species only occurred between 10 and 50km from source. It is possible that some of the sites with just four or five species had experienced some form of natural or man-induced stress prior to sampling and may be capable of supporting more species. The most species rich sites have been labelled 1-9 and are identified in a footnote to Fig.2.2. They include a number of the most famous chalk streams in southern England (R. Frome in Dorset, R.Test, R.Itchen and R.Lambourn) and other rivers (R.Crane/Moors River) recognised for the high species richness of their macroinvertebrate assemblages (Wright et al.1988).
2.4. Longitudinal occurrence of species
In order to investigate the basis of this variation in site richness, it is necessary to examine the longitudinal occurrence of the individual species. In Table 2 3 the 42 sites have been placed into a series of eight categories representing distance from source. The distance categories are <5, <10, <15, <20, <30, <40, < 60 and <100km from source. The total number of sites in each distance category is given at the head of the table and this is followed by information on the longitudinal occurrence of each species
Within the Bactidae, Baths rhodani, B.vernus and also Centroptilum luteolum were common or frequent in all distance categories. Several other species, including B.muticus, B. niger and B. scambus group were also widely distributed, although less frequent at sites <10 km from source. Additional species in the genus Baths (B. atrebatinus, buceratus and digitalus) appeared most consistently in the middle and lower reaches of chalk streams, as did C. pennulatum, Cloeon dipterum and Procloeon bifidum.
In the Heptageniidac, whereas both Heptagenia sulphurea and Ecdyonurus p. were widely distributed, Rhithrogena sp. was never recorded more than 15 km from source.
Similarly, within the Leptophlebiidae there was evidence of spatial separation of species. Leptophlebia vespertina is more characteristic of ponds, lakes and slow- flowing streams and hence the single record in the lower reaches of a chalk stream was to be not entirely unexpected. Whereas Paraleptophlebia submarginata is widespread in chalk streams, P.citwta, though occasionally recorded, is more frequent in other river systems. In contrast, P.werneri, which has R.DB3 status, is normally found in winterbournes and the upper perennial sections of chalk streams. Although the single R1VPACS site rccord was at 5.1 km from source on the R.Crane, the species was also recorded at a further site 3 5 km from source on the R. Crane which dried out in summer and was rejected as a R1VPACS site. Habrophlebia fusca is also most frequently encountered in small streams.
Both the Ephemeridae and Ephemerellidae were represented by single species (Ephemera danica and Ephemerella ignita respectively), which were widely distributed along the length of southern chalk streams
Finally, the four taxa within the Caenidae included two species (Brachycentrus harrisella and Caenis horaria) typical of silty rivers which were restricted to downstream locations and two others (C. rivulorum and C. luctiosa group) which were more widespread in their occurrence.
Table 2.3 Longitudinal occurrence of mayfly larvae at 42 chalk stream sites
Distancefrom Source (km) <5 <10 <15 <20 <30 <40 <60 <100 Number of sites in cute or 6 5 8 5 5 5 4 4
3 I Baetis atrebatinus - I 2 4 4 Baths buceratus I I I 1 2 - - Baetis digttatus - 1 Baetis muticus 1 2 3 1 1 5 4 - Baetis mger 1 I 5 1 3 4 2 Baetis rhodani 6 5 8 5 5 5 4 3 3 3 Baetis vernus 5 5 8 4 5 4 &Jetts scambus group I I 5 5 5 5 4 4 Centroptilum luteolum 2 3 7 3 3 3 2 2 Centroptilum pennulaturn - I 2 2 1 1 Cloeon dipterum - 1 1 _ Procloeon hifidurn - 1 I Rhithrogena sp. I 2 2 3 1 Heptagenia sulphurea 2 2 5 2 4 5 Ecdyonurus sp. I I I I Leptophlebia vespertina - 1 1 Paraleptophlebia cincta 1 1 I Paraleptophlebia submarginata 2 3 3 2 3 5 Paraleptophlebia werneri I - Habrophlebia fusca I I Ephemera danica 4 3 7 4 4 3 4 2 Ephemerella ignita 6 5 8 5 5 5 4 4 Brachycercus harrisella I I Caenis horaria - 1 1 Caenis rivulorum 3 2 5 2 4 4 3 Caenis luctuosa group 2 I 3 5 4 3 3 4
2.5. Abundance of families at chalk stream sites
The species-level data for mayflies at all R1VPACS sites is simply presence/absence data with no indication of whether the individual species are abundant, common or scarce at each site. Each 3-minute R1VPACS sample was obtained by pond-netting all available habitats, with the time roughly allocated in proportion to the area of each habitat. The resulting sample was therefore effort-dependant, and whilst it was NOT a quantitative sample, it did include crude information on whether taxa were common or scarce. In view of the potential value of this information, it was decided to estimate the abundance of every family of macroinvertebrate in each sample in terms of log categories of abundance (i e 1=less than 10 individuals per sample; 2=<100, 3=<1,000; 4= Baetidae were present at almost all sites in each of spring, summer and autumn. They were also a dominant element of the mayfly fauna in most seasons, particularly in summer when over 30 sites recorded log category 3 abundance (i.e. between 100 and 999 specimens per 3-minute sample). The Heptageniidae were dominated by a single species, fleptagema sulphurea and summer emergence dictated that densities of larvae would be lowest at this time. The Leptophlebiidae were also dominated by a single species (Paraleptophlebia submarginata) and once again, early summer emergence resulted in lower densities of larvae at this time before the next generation appeared in autumn. Ephemera danica, the only species of Ephemeridae recorded at the 42 chalk stream sites, has a two year life cycle and therefore, apart from the emergence of some larvae in the second half of May resulting in lower abundances in summer, the changes in abundance were relatively damped. Once again, the Ephemerellidae were represented by a single species, the ubiquitous Ephemerella ignita. Although this species was recorded at every site (Table 2.1), the life cycle resulted in most sites having their highest abundances of this important member of the mayfly fauna in summer. Finally, the Caenidae, with four taxa included in the data-set (plus the recently recorded C.pusilla), appeared to be progressively less abundant from spring to autumn. The number of sites at which they occurred also decreased with season. Standard RIVPACS samples, as collected and processed by the Environment Agency, frequently have log, abundance categories ascribed to each I3MWP family, including the six families of Ephemeroptera featured above. The information presented in Fig. 2.3 provides a record of the range of abundance categories observed over 42 different chalk stream sites in three seasons. Previously unsampled chalk stream sites which are examined in a given season could exhibit any of the results shown in the figure. However, at high quality sites, some results are more likely than others, and for example, there are very few chalk stream sites in Fig. 2.3 in which the Baetidae are not abundant in summer. Hence, used with caution, this Figure may offer some early clues on whether newly sampled sites are poorly represented in terms of their mayfly larvae Note, however, that the use of full R1V1ACS predictions at the family abundance level will provide more detailed predictions tailored to the specific attributes of new sites. 1(1 I•111 a I= =I On MO MI MN I= NM MN MIS 231 22- 20 23 E a I yl 10; 10 1 1°01 0 o c,, Tr 0 c., .1 0 N V' 0 31 36 a. al al at: 231 g. P 5 id 10 101 10 0 0 0 N vt 0 CV .1 itt •ct 0 31 30 21 20 101 10 01 0 0 o N Log abundance Baendae Heptagentidae Leptophlebndae Ephemendae Ephemerellidae Caenidae Fig 2.3. The log category abundance data for each mayfly family at 42 chalk stream sites in spring, summer and autumn. The '0' log abundance category (unshaded) indicates the number of sites at which the family was absent in a given season. Log category 1 = <10 individuals per sample; 2 = <100; 3 = < 1,000; 4 = <10,000. IMMEM OM MI - 01= I= NM MN a NM OM a a a a a IM r%1 - 3. NOTES ON THE LIFE CYCLES OF EACH SPECIES Despite the fact that a very substantial number of studies have been conducted on the life cycles of mayflies in Europe, our knowledge is far from complete This is, in part, due to the fact that in some mayflies the life cycle shows flexibility, depending on the environmental conditions. In addition, where a species is capable of having several generations in a year, interpretation of the life cycle can be problematic when based on field observations alone. Hence, many workers in this field emphasize the importance of both field and laboratory studies (egg hatching and larval growth) for the correct interpretation of the life cycles (Brittain, 1982, Elliott et al. 1988) Elliott and Humpesch (1983) and Elliott el a/. (1988) have collated the extensive European literature on the flight periods and life cycles of mayflies found in Britain. Information from these two publications is reproduced in Table 3.1 for the 26 taxa being considered in this report. Note that the flight periods include observations from many varied streams and rivers within Europe and so, flight periods in southern chalk streams may be more restricted in some cases. An example is Ephemera danica, which is normally observed in May/June, although emergence at other times of the year is not unknown. Elliott et a/. (1988) classify the life cycles of the British species into five broad categories, which are listed and defined in a footnote to Table 3.1. Groups IA and 1B are both univoltine (one generation per year) but whereas IA ovenvinters in the egg stage, IB overwinters in the larval stage. Groups 2A and 2B are bivoltine (two generations) or multivoltine (more than two generations per year) but again, 2A overwinters in the egg and 213overwinters in the larval stage. Finally, group 3 has one generation every two years (every three years in some northern latitudes). Note that more than one category is offered for some species (where life cycle varies with river type or geographical location) and that a question mark is used where uncertainty remains. The final column in Table 3.1 is reserved for additional notes on some species whose flight periods/life cycles have been studied in southern chalk streams. The Baetidae in particular, display a considerable degree of life cycle flexibility throughout their distributional range (Brittain, 1982). For example, Haetis rhodani may have a single overwintering generation in northern Europe and in mountainous areas further south. However, a winter generation followed by a summer generation is typical of less extreme conditions and in more southerly locations there may be additional summer generations. Elliott et al (1988) point out that the combination of the long flight period, the long period when eggs are present and the multivoltine life cycle arc the major factors which enable this species to dominate the mayfly fauna of most streams and rivers in Britain Table 3.1 indicates that all members of the Bactidae are capable of having two or more generations per year (Groups 2A & 2B). Clearly, those with overwintering larvae II (Group 213) will have greater capacity for early spring emergence than those which overwinter in the egg stage (Group 2A). Note that B.vernus & B. scamhus group, Table 31 Information on the flight periods and life cycles of mayflies abstracted from Elliott and Humpesch, (1983) and Elliott el at (1988) respectively, together with additional observations made in southern chalk streams. Notes Species Flight Period Life Cycle Baetis atrebatinus may - Oct 213(9) Baetis buceratus Apr - Oct 2B (The(is digitatus May - Scp Baetis mutacus Apr - Oct 28 Baetis niger Apr - Oct 28 Multivoltine (Welton et at 1982) Baths rhodans Jan - Dec 28 Pre-ernergent larvae Mar - Dec' Pre-emergent larvae April -Dec' Baetts vernus Apr - Oct 1A/2A Baths scambus group Feb - Nov 1A/2A Pre-emergent larvae May - Dec' Centropttlum Iuteolum Apr - Nov 28 2 or possibly 3 generations (Welton et at 1982) Centroptilum pennulatunt May - Oct 2A(?) Cloeon dtptertan May - Oct 28 Procloeon bifidum Apr - Oct 2A(?) Rhithrogena sp. Mar - Sep I B Ileptagenta sulphurea May - Oct I B Ecdyonurus sp. Mar - Oct I B Leptophlebia vespertina Apr - Aug 1B Paratcptophlebia cincta May - Aug IANIB(?) et at 1982) Paraleptophlebia submarginata Apr - Jul IB Univoltine (Welton Paraleptophlebia werneri May - Jun I A(?) Univoltine in Winterbournes Habrophlebta fusca May - Sep 1B Ephemera danica Apr - Nov I B/3 Semivoltine (Wright et al. 1981) Apr - Nov' Ephemerella ignita Apr - Sept I A/ I B/2B Pre-emergent larvae See also Bass (1976) & Welton et at (1982) Brachycercus harrisella - July -7 1A(7) Caenis horaria May - Sep I B/2B Caenis rivulorum Apr - Sep I B Univoltinc (Welton et at 1982) Pre-emergent larvae May -June' Caenis luctuosa group May - Sep IB/2B Group IA. Univoltine (onc generation per ycar), ovenvintering in egg stage. Group 1B. Univoltine, overwintering in larval stage. Group 2k Bivoltine (two generations per year) or multivollinc (more than two generations per year), overwintcring in egg stage. Group 2B. Bivoltinc or multivoltine, overwintering in larval stage. Group 3. Sentivoltine (one generation every two years or even every three years). Data on the occurrence of pre-emergent larvae taken front University of Reading (1978) 14 which have overwintering eggs, (and also C pennulantm & Procloeon hifidum which are believed to have overwintering eggs), are less frequently encountered in spring samples in relation to those collected in summer and autumn (see Table 2.2). Welton et at(1982) studied the growth and production of five species of mayflies in a recirculating channel fed by borehole water whose chemistry was similar to local chalk streams in Dorset. The most abundant species was B.rhodani, and they were able to recognise five cohorts through the year. Their study also included C. luteolum, which had two and possibly three cohorts per year. In a separate study conducted at seven chalk stream sites during the mid-1970's (University of Reading, 1978) 5-minute pond-net samples were collected each week and pre-emergent larvae (with darkened wing-buds) were removed to document emergence patterns and attempt to recognise discrete generations. Pre-emergent larvae belonging to B.rhodani, B.vernus and B.scambus group began to appear at various times in spring (Table 3.1) and were sometimes recorded into December. In all cases there was convincing evidence of at least two generations. In general, those Heptageniidac and Leptophlebiidae represented in Table 3.1 appear to have simple univoltine life cycles with overwintering larvae in most cases. However Paraleptophlehia werneri, which is believed to overwinter in the egg stage, must adopt this strategy in winterbourne streams, which dry out in summer leaving the resistant eggs within the stream bed. After the springs break in winter, the eggs can then hatch in early spring and rapid larval growth allows emergence to take place before the stream dries out once more. In southern chalk streams, Ephemera danica has been shown to have a two year life cycle (Wright et at 1981), with emergence in late May/ early June Whereas Ephemerella ignita is univoltine with overwintering eggs in many streams in Britain, larvae occur throughout the year in southern chalk streams (Bass, 1976). University of Reading (1978) recorded pre-emergent larvae from April to November and Kite (1962) observed adults on the wing in all months of the year in southern England. The reason for these differences is related to egg hatching. Elliott et at (1988) demonstrated that in cool streams, a small proportion of eggs do hatch through the autumn and winter, but the main egg hatching period is between March and May, resulting in emergence between April and September. Chalk streams remain relatively warm through' the autumn and winter months, allowing a higher percentage of eggs to hatch. Nevertheless, the highest densities of larvae recorded in small chalk streams were in April (Welton et a/. 1982) or May (Bass 1976) and the peak densities of pre- emergent larvae in the seven chalk streams examined by University of Reading (1978) were observed between May and July. Information is limited for the Caenidac. The single generation of Caents rivulorum studied in the recirculating channel (Welton et at 1982) started to emerge in May, and University of Reading (1978) only recorded pre-emergent nymphs in May and June. The seasonal occurrence of C. luctuosa group (Table 2 2) in which specimens were recorded at most sites in summer suggests somewhat later emergence than C. rivulorum 15 MO MENNIO a a Ma MIMIMSIMMIMIMMIMEMMOIM 4. FACTORS AFFECTING THE DISTRIBUTION AND ABUNDANCE OF EPIIEMEROPTERA 4.1. Habitats, habits and feeding behaviour of the larvae Elliott el al. (1988) provide a summary of the habitats, habits and feeding behaviour of the British Ephemeroptera. More recently, Wright el at (1993) undertook a survey of macroinvertebrate-habitat associations at 76 lowland running-water sites in England and Wales A large number of different habitats were sampled within three broad categories. non-macrophyte, submerged macrophyte and emergent macrophyte. Early analyses, based on these categories and using mayflies at family level only, demonstrated statistically significant positive associations between submerged macrophytes (Baetidae & Ephemerellidae), emergent macrophytes (Leptophlebiidae), non-macrophytes (Eleptageniidae & Ephemeridae) and no significant associations for Caenidae (Wright ci al 1994). The habitat preferences of the individual species of mayflies in this study (Wright et al. 1993) are given in italics in Table 4.1. All other details on habit and feeding behaviour are taken directly from Elliott ei at (1988). The descriptions of 'habit' need brief explanation. 'Swimmers', including many Baetidae, have a torpedo-shaped body whereas 'clingers' (e.g Heptageniidae) are dorso-ventrally flattened and have large curved tarsal claws. Sprawlers (e.g Caenidae) are poor swimmers and live in the surface layers of fine sediment or on the surface of macrophytes. 'Climbers' (eg Centropnlum spp ) are adapted to life amongst the stems of macrophytes whilst burrowers such as Paraleplophlebia spp. and Ephemera danica live in river sediments Larvae of all thc British species of mayflies are herbivores and feed on detritus or periphyton (algae and associated material on macrophytes and non-macrophyte substrata). 'Scrapers' exploit periphyton, 'collector gatherers' eat fine deposited detritus, whilst 'collector filterers' feed on fine detritus in suspension Note that the diet of larvae is not fixed but may vary with the stage of development, time of year and the habitat/river in which the species occurs. Further information on both habits and feeding types may be found in Elliott et at 1988. Table 4.1 Running-water habitats (from Wright et at 1993), together with habits and feeding behaviour (from Elliott et aL 1988) for thc mayflies found in southern chalk streams. See text for further explanation Species Running water habitat Habit Feeding Behaviour Baetis atrebatinus Submerged & emergent Swimmer Scraper & collector- gatherer macrophytes Baths bucernms Submerged macrophoes Swimmer Scraper & collector- gatherer Baths digitatus Submerged macrophytes Swinuned climber Scraper & collector- gatherer Baths muncus Submerged & emergent Burrower Scraper-gathererAscraper") macrophytes Lketis niger Submetxed & emergenl Swimmer/climber Scraper and -gatherer macrophprs 17 Scr aper and collector Baths rl«xlani Submerged & emergent Swimmer - gather er macrophytes plus non. macrophytes Scraper and collector-gatherer Baths vemus Submetged & emergent Swimmer macrophytes Scraper and collector-gatherer Baths scambus group Submerged & emergent Swimmer/Climber macrophytes Collector-gatherer Centroptilum luteolum Emergent macrophytes Swimmer/Climber Swimmer/Climber Collector-gatherer Centroptilum pennulatum Emergent macrophytes Swimmer/CI imber Collector-gatherer Cloeon dipterion Emergent macrophyles Swimmer/CI imber Collector-gatherer Procloeon bifidum Emergent & submerged macrophytes plus non- macrophytes Clinger Scraper and collector-gatherer Rhithrogena sp Non-macrophytes Clinger & swimmer Scraper and collector-gatherer Heptagenia sulphurea Submerged macrophytes Clinger & swimmer Scraper and collector-gatherer Ecdyonunts sp. Non-macrophytes Sprawler & climber Collector-gatherer Leptophlebia vespernna Running water (pools/margins and on macrOphytes) Burrower Collector-gatherer Pamleptophlebia moo Submerged & emergent macrophytes Burrower Collector-gatherer Paraleptophlebia submarginata Submerged macrophytes Burrower Collector-gatherer Paraleptophlebta wernen Emergent macrophytes Sprawler & climber Collector-gatherer Habrophlebia fusca Emergent & submerged macrophytes plus non- macrophytes Burrower Collector-filterer Ephemera danica Non-macrophytes plus submerged &emergent